Ray.Di | A dependency injection framework for PHP | Dependency Injection library

The jump logic should be inside the Update() method since we want the height to be calculated every frame.

Well your issue is pretty simple:

You are checking the distance between the player position and the ray direction!

Of course this makes absolutely no sense since it basically equals the Camera.main.transform.forward. If you use a normalized direction like a position it will basically be something around the Unity origin at a distance of 1. While your player could be positioned just anywhere. You want to check the distance between player and hitInfo.point!

When you first hit the object, you could remove or disable it's collider component. This should prevent it from being hit by more Raycasts.

On first hit, something like:

I believe “scope” is by default per request which would explain that each submit gets is own service.

“Doing stuff right” is of course to some extend a matter of opinion. But my opinion would clearly be that I would avoid server-side session to avoid problems with scaling to more than one instance. There are also ways to support shared state, but this is difficult. To me singletons are not a code smell either, but they have their own problems.

Your problem might be solved by storing whatever state you need in the browser either in a cookie or localStorage. Your service would then have request scope, but it would read user state from browser causing “user scope” for the data. (But don’t rely on browser state to persist and remember it is modifiable to the user.)

I think I sort of tracked down what the problem was, thanks to Retired Ninja in the comment section. I reduced the code to just a few lines, forgetting about the mesh system and all that stuff. Apparently, when I grab the window, the main thread seems to sleep until I am done moving it, and in the ray-trace algorithm I am instantiating new threads, so when the main one sleeps, most of the times it doesn't await (aka "join()") for them, and because I am passing pointers to those threads, those seem to fill up the memory and crash. I have added a system to also sleep the working threads while the main one is doing so and it works like a charm!. Now I gotta see how to do it in cuda XD. Nevertheless , thank you all!

All spheres are white most likely because ClosestSphere.Hit in GetRayColor function is always true.

I think the problem is in your IntersectSceneSpheres function.

In CPU code you return HitAnything which is defaulted to false. At the same time in the fragment shader you return struct ClosestRecord that remains uninitialized if nothing was hit.

Explicitly adding ClosestRecord.Hit = HitAnything; in the end of IntersectSceneSpheres function should fix it

I found my error, and in case someone needs this int the future I'll try to describe it here. So basically, when rendering an infinite cylinder we try to get the closest intersection and we render it. For a finite cylinder that we cut, it doesn't work. If the distance between the intersection and the center of the cylinder is higher than the length of the cylinder, then we do not consider an intersection. Even though there is no intersection at the closest intersection (which is too high), there is one behind it which might be in the range of the height of the cylinder. And this is the intersection that we need to get the other side of our cylinder.

Maybe some pseudo code would help

One of your parameters is named ray, so in the function body any references to ray refer to that parameter, not the class name the function is defined in.

The solution is simple: call your parameter something else.

Properties of your Sphere class are readonly, you should do them like below ({get;set;}):